The deployment of application topologies within an enterprise is both error-prone and time-consuming. Moreover, the number of errors during deployment and the time of deployment increase as the complexity of an application topology increases. In addition, an error in the deployment of an application topology may result in significant loss in productivity of the enterprise. For example, the error may result in an application that is critical to the enterprise being unavailable. The application topology of an enterprise is typically defined by a system administrator. The system administrator may decide the interconnection between the computers and decide which applications should execute on which server and client computers. The system administrator for a large enterprise may define various domains for the computers of the enterprise. For example, the system administrator may define a domain for each location of the enterprise and specify that each of the domains has a trust relationship with the other domains. The system administrator may also specify the applications that should be installed on each computer. The applications that should be installed on a computer depend upon the roles assigned to that computer. For example, the system administrator may assign one computer the role of domain controller, another computer the role of domain name service for the computers of the domain, another computer the role of certificate authority for the domain, and so on. Other roles may include electronic mail server, network meeting server, and so on. To provide the services of these roles, each of the computers needs to have installed the applications (i.e., the software components) necessary to perform those roles.
As the application topology of an enterprise evolves over time, it can even be difficult for a system administrator to determine what is the current application topology. As a result, it can be difficult for the system administrator to know the consequences of a change to the application topology. For example, a system administrator may inadvertently move a critical application from one computer to another computer that may not have the needed processing power or that may not be connected to all the computers that need the services of that application. Thus, the system administrator may only fully understand the consequences of a change to an application topology after implementation. Any unintended consequences, however, may result in failure of one or more applications of the network or even failure of the network itself.